Sp8 exhibits reciprocal induction with Fgf8 but has an opposing effect on anterior-posterior cortical area patterning
© Sahara et al.; licensee BioMed Central Ltd. 2007
Received: 16 March 2007
Accepted: 17 May 2007
Published: 17 May 2007
Telencephalic patterning centers, defined by the discrete expression domains of distinct morphogens, Fgf s in the commissural plate (CoP), Wnt s and Bmp s in the cortical hem, and a ventral domain of Sonic hedgehog (Shh), are postulated to establish during development the initial patterning of the telencepahlon, including the neocortex. We show that the expression patterns of Sp5, Sp8, and Sp9, members of the Sp8-like family that are homologues of Drosophila buttonhead, correlate during early embryonic development with these three telencephalic patterning centers. To study potential functional relationships, we focused on Sp8, because it is transiently expressed in the CoP coincident with the expression of Fgf8, a morphogen implicated in area patterning of the neocortex. We also show that Sp8 is expressed in cortical progenitors in a high to low anterior-medial to posterior-lateral gradient across the ventricular zone. We used in utero electroporation of full-length and chimeric expression constructs to perform gain-of-function and loss-of-function studies of interactions between Sp8 and Fgf8 and their roles in cortical area patterning. We show that Fgf8 and Sp8 exhibit reciprocal induction in vivo in the embryonic telencephalon. Sp8 also induces downstream targets of Fgf8, including ETS transcription factors. In vitro assays show that Sp8 binds Fgf8 regulatory elements and is a direct transcriptional activator of Fgf8. We also show that Sp8 induction of Fgf8 is repressed by Emx2 in vitro, suggesting a mechanism to limit Fgf8 expression to the CoP. In vivo expression of a dominant negative Sp8 in the CoP indicates that Sp8 maintains expression of Fgf8 and also its effect on area patterning. Ectopic expression of Sp8 in anterior or posterior cortical poles induces significant anterior or posterior shifts in area patterning, respectively, paralleled by changes in expression of gene markers of positional identity. These effects of Sp8 on area patterning oppose those induced by ectopic expression of Fgf8, suggesting that in parallel to regulating Fgf8 expression, Sp8 also activates a distinct signaling pathway for cortical area patterning. In summary, Sp8 and Fgf8 robustly induce one another, and may act to balance the anterior-posterior area patterning of the cortex.
Many homeotic genes originally identified in Drosophila have important roles in vertebrate development. Buttonhead (btd), empty spiracles (ems), and orthodenticles (otd) were identified through mutagenesis screens as genes required for anterior head development in Drosophila [1–3]. Vertebrate homologues of ems and otd, which encode the Emx and Otx families of transcription factors (TFs), respectively, have been studied extensively, particularly their roles in mammalian forebrain development [4, 5]. Vertebrate homologues of btd, a member of the Sp1-Zn finger family of TFs, have also been identified [6–8], but in contrast to homologues of ems and otd, they have been studied relatively little, particularly their roles in brain development.
Nine genes have been identified in mammals as btd homologues and the proteins they encode share btd and Zn finger domains. Based on the domain structures of their protein products, these nine genes are divided into a Sp1-like family (Sp1-4) and a Sp8-like family (Sp5–9). However, because Sp8 can rescue the defects in head development in Drosophila btd mutants, whereas Sp1 cannot , Sp8 and its family are viewed as the functional homologues of Drosophila btd. Mice deficient for either Sp5 or Sp8 have been reported. Sp5 knockout mice reportedly have no overt phenotypes . In contrast, analyses of Sp8 knockout mice [9, 11], complemented by retrovirus-mediated overexpression studies in embryonic chicks using dominant negative and active forms of Sp8 , indicate that Sp8 is required for limb and head formation.
A prominent role for Sp8 in limb development is to maintain the expression of several morphogens, including fibroblast growth factor (Fgf)8, bone morphogenic protein (Bmp)4, and Sonic hedgehog (Shh) [9, 11]. Because these morphogens have prominent roles in forebrain development, we examined the expression of the Sp8-like family members in the developing mouse forebrain to determine their potential for roles in forebrain patterning, particularly the patterning of the neocortex into 'areas', which are anatomically and functionally distinct divisions with specialized functions in sensory perception, motor performance, learning and memory.
The genetic mechanisms that control cortical area patterning, that is, arealization, have come under intense study in the past few years, but remain sketchy. Current evidence indicates that morphogens secreted by signaling centers located at the perimeter of the developing dorsal telencephalon (dTel) establish gradients of TFs across the dTel ventricular zone (VZ), which in turn determine the area fate of cortical progenitors and their progeny [12–17]. These morphogens include Fgf8 expressed by the anterior neural ridge (ANR), which later becomes the commissural plate (CoP), located at the anterior midline of the dTel, and also members of the wingless-int (Wnt s) and Bmp families expressed by the cortical hem and choroid plexus epithelium (CPe), located at the dorsal midline of more posterior dTel. In addition, the morphogen Sonic hedgehog (Shh) is expressed in the forebrain, but in ventral structures, including the prechordal plate and the medial ganglionic eminence (MGE).
Here we describe that each of these three major signaling centers in mouse forebrain is uniquely associated with the discrete expression patterns of a member of the Sp8-like btd family, Sp5, Sp8 and Sp9. These associations are suggestive of functional interactions between the Sp family members and the respective morphogens. To address this issue, we focused on functions of Sp8, which at early stages of cortical development is expressed in association with the expression domain of Fgf8 in the ANR/CoP. We employed both gain and loss of function strategies using in utero electroporation to manipulate gene function in a localized manner, a technique similar to that recently used by others to show roles for Fgf8 in cortical area patterning [18, 19]. We complemented these in vivo studies with in vitro assays of inductive pathways and interactions. These approaches allowed us to overcome the early embryonic lethality and forebrain defects exhibited by Sp8 knockout mice, including the failure of the anterior neural tube to close and the cerebral cortex to develop, that preclude an analysis of Sp8 in cortical patterning [9, 11].
We show that Sp8 and Fgf8 can robustly induce one another, and that the induction of Fgf8 by Sp8 is repressed by the homeodomain TF, Emx2, which itself has been implicated in controlling area patterning [19–22]. The reciprocal induction between Sp8 and Fgf8 exhibits significant specificity among family members. In addition, we show that Sp8 affects cortical area patterning, but in a manner that opposes that of Fgf8, raising the possibility that Sp8 influences area patterning not only by regulating Fgf8 and its signaling pathway, but also through a distinct mechanism. Our findings show that Sp8 interacts with Fgf8 to balance the anterior-posterior (A-P) area patterning of the cortex.
Forebrain expression of the Sp8-like family, mouse btd homologues, and correlation with signaling centers in the forebrain
Between E9.5 and E11.5, expression of Sp5 is highest in the medial and posterior dTel, including the cortical hem and CPe, and exhibits a very steep gradient with the cortical VZ, quickly declining from high to low levels within more posterior and medial dTel, resulting in very low levels in more anterior and lateral dTel (Figure 1a, b, d, e, g, h). The pattern of Sp5 expression is considerably more extensive than the cortical hem/CPe, but most of the caudomedial part of the Sp5 expression domain contains the Wnt2b expression domain (Figure 1c, f, i), a marker for the cortical hem . By E13.5, Sp5 expression declines to non-detectable levels in the cortex, but persists in the hem and CPe (Figure 1). Sp5 is also expressed at lower levels in the mantle of the MGE.
At E10.5, Sp9 is modestly expressed in the mantle zone of the ganglionic eminence (GE) (Figure 2a, b) and, therefore, likely in postmitotic neurons. At E11.5 (Figure 2d, e), Sp9 is robustly expressed in the MGE, throughout the mantle zone and likely also in the subventricular zone (SVZ) of the MGE. At this age, the MGE mantle includes postmitotic GABAergic interneurons that migrate into and populate the cortex , and the SVZ contains proliferating cells. Sp9 is also expressed robustly in the SVZ of the lateral GE (LGE). Sp9 expression overlaps with Shh expression in the MGE mantle. However, elsewhere in ventral telencephalon, Shh is predominantly expressed in largely non-overlapping, complementary patterns with Sp9 (Figure 2).
From E9.5 through E13.5, Sp8 is expressed in a high to low medial to lateral gradient across the dTel VZ, and is expressed across the A-P axis of dTel, with a shallow high to low A-P gradient; by E10.5, Sp8 is expressed across the entire A-P axis. The Sp8 domain is associated with the Fgf8 expression domain in the CoP (Figure 3b, c). The correlation of the Sp8 domain with the Fgf8 expression domain in the CoP is dynamic over the brief time window of Fgf8 expression (Figure 3b, c, e, f, h, i). At E9.5 and E10.5, the Sp8 and Fgf8 domains overlap, although the Sp8 domain is more expansive and extends laterally and posteriorly from the CoP through the dTel VZ (Figure 3b, c, j, k). At E11.5, Sp8 is robustly expressed, but it is beginning to segregate from the Fgf8 domain at the posterior domain, with Fgf8 largely limited to the midline CoP, and Sp8 largely excluded from it and expressed in a high to low gradient extending laterally from the midline (Figure 3e, f). At E13.5, Sp8 expression in the dTel VZ is still present but substantially diminished, and Sp8 has become excluded from the increasingly restricted Fgf8 domain in the CoP (Figure 3h, i).
Sp8 is also expressed in the SVZ of the LGE, a source of interneurons that migrate into the olfactory bulbs, and at low levels in the mantle of the LGE and MGE. Sp8 has been shown to regulate the specification, survival and migration of olfactory interneurons .
Sp8 and Fgf8 exhibit reciprocal induction
To address whether Sp8 reciprocally induces Fgf8, we ectopically transfected expression vectors for Sp8, or to assess specificity, Sp5 and Sp9. We found that ectopic expression of Sp8 robustly induces Fgf8 expression (n = 9 of 10, Figure 4j). Ectopic expression of Sp9 only modestly induces Fgf8 (n = 8 of 9, Figure 4l), whereas Sp5 does not induce detectable levels of Fgf8 (n = 2 of 2, Figure 4n). These findings show that Sp8 can induce Fgf8, and, therefore, that Sp8 and Fgf8 form a reciprocal inductive loop.
To address whether Sp8 is a transcriptional activator of Fgf8, we used a strategy of fusing the btd and Zn finger functional binding domains of Sp8 to a VP16 domain, which is a potent transcriptional activator . This chimeric construct acts as a dominant active form of Sp8 that can induce Fgf8 in the developing chick limb . We found that ectopic expression of the Sp8-VP16 construct also induces Fgf8 (n = 12 of 14, Figure 4p ), indicating that Sp8 is a transcriptional activator of Fgf8. The VP16 construct alone (n = 6 of 6, data not shown) does not induce Fgf8. In summary, we demonstrate that Fgf8 and Sp8 exhibit reciprocal induction in vivo in the embryonic telencephalon. Furthermore, our findings show that Sp8 regulates transcription of Fgf8.
Fgf17 is expressed within the CoP, albeit in a broader domain than Fgf8, and has a similar effect as Fgf8 on A-P cortical patterning . Therefore, we also performed in utero electroporations to test whether Sp8 and Fgf17 exhibit reciprocal induction. However, we did not detect reciprocal induction (n = 3 of 3 cases, data not shown), suggesting that Sp8 is not directly relevant to Fgf17 signaling. Thus, Sp8 and Fgf17 do not induce one another.
In conclusion, the reciprocal induction between Sp8 and Fgf8 exhibits a specificity that does not extend significantly to related Fgf and Sp8-like family members.
Sp8 activates the Fgf8 signaling pathway
Sp8 binds the Fgf8 promoter and activates Fgf8 expression
To show Sp8 binding to Fgf8 promoter sequences, we first carried out a gel retardation assay. This 585 bp fragment of the Fgf8 promoter region was labeled with P32 and incubated with lysate from reticulocytes expressing the Sp8 protein under the control of a T7 promoter, or control lysate. We detected two different bands in the Sp8 lysate but not in the control lysate (Figure 6b). To corroborate the specificity of these results, we performed a competition experiment by repeating the gel retardation assay as above, but in the presence of an excess of the 585 bp DNA fragment not labeled with P32. In this competition assay, the bands were absent, indicating that Sp8 protein directly binds to these fragments of the Fgf8 promoter region.
As a complementary approach, we tested Sp8 binding to the Fgf8 promoter sequences by a pull-down assay. Lysates from cells transfected with hemagglutinin (HA)-tagged Sp8 or control vector were incubated with a biotinylated oligonucleotide corresponding to the putative Sp8 binding site, or with one mutated at the core region of the Sp1-recognition sequences. We found co-precipitation of Sp8 with the wild-type oligonucleotide but not with the mutated version (Figure 6c). Thus, these data indicate that Sp8 recognizes and binds to this region of the Fgf8 promoter defined by sequence similarity to Sp1 binding sites.
To further investigate if Sp8 controls Fgf8 expression through binding this promoter region, we used a luciferase reporter assay (Figure 6d). Control or test vectors containing Sp8, Sp9, or Emx2 were transfected independently or in combinations into C3H10T1/2 cells also transfected with an Fgf8 reporter construct consisting of the same 585 bp promoter element for Fgf8 used in the gel retardation assay, and a firefly luciferase reporter vector . Transfection of an Sp8 expression vector robustly induces expression of the Fgf8 reporter construct, whereas Sp9 has only a modest inductive effect.
Noggin, a Bmp inhibitor, has been recently reported to induce Fgf8 expression when overexpressed in forebrain . Therefore, we tested whether a suppression of Bmp signaling could contribute to Sp8-mediated Fgf8 upregulation. However, we found that manipulation of Bmp activity has no effect on Sp8 transcriptional activation of Fgf8 in the in vitro reporter assay (Additional file 1).
Taken together these findings show that Sp8 binds to a proximal promoter region of the Fgf8 gene of Fgf8, and that this binding induces Fgf8 expression.
Sp8 induction of Fgf8 is repressed by Emx2
As described above, Sp8 is expressed in the CoP coincident with Fgf8. However, Sp8 is also expressed outside of the CoP, within the VZ of dTel (neocortex), locations where Fgf8 is not expressed, suggesting that factors expressed in the cortical VZ repress the ability of Sp8 to induce Fgf8 and restrict its expression to the CoP. Emx2 is a candidate repressor because the domain of Fgf8 expression in the CoP is broader in Emx2 null mice, and ectopic Emx2 expression in the CoP in slice culture suppresses Fgf8 expression .
Therefore, we tested whether Emx2 can repress Sp8 induction of Fgf8 in a luciferase reporter assay. An Emx2 construct transfected alone has no effect on expression of the Fgf8 reporter, and is indistinguishable from a control empty vector alone. However, when Emx2 is co-transfected with Sp8, Emx2 represses Sp8 induction of the Fgf8 reporter construct. Emx2 also represses the modest induction of Fgf8 exhibited by Sp9.
In summary, these findings are consistent with our in vivo analyses, and provide a molecular mechanism for them, demonstrating that Sp8 binds Fgf8 regulatory elements and is likely a direct transcriptional activator of Fgf8. Furthermore, they show that Emx2 can act as a repressor of Sp8 induction of Fgf8.
Sp8 is required to maintain Fgf8 expression in vivo
These results were also confirmed by the expression of Pea3, a member of the Pea3 ETS TF subfamily that is rigidly regulated by Fgf8 [31, 32]. We found that transfection of the dominant negative Sp8-engrailed construct into the CoP results in a substantially diminished expression of Pea3 (n = 7 of 8, Figure 7f), whereas the control engrailed construct has no effect (n = 9 of 9, Figure 7e). These results demonstrate that Sp8 is required to maintain Fgf8 expression within the CoP in vivo, as well as expression of its targets.
Maintained expression of Fgf8 by Sp8 is required for proper cortical area patterning
If Sp8 is required to maintain Fgf8 expression and activation of its signaling pathway, as our results above show, then electroporation of the dominant negative form of Sp8 (Sp8-engrailed) into anterior dTel should affect cortical area patterning by diminishing Fgf8 signaling. We predict that the effect on area patterning should resemble the anterior area shifts produced by ectopic expression in anterior dTel of the extracellular domain of Fgf receptor 3c (sFGFR3c), which is interpreted to be due to sFGFR3c binding and suppressing the action of endogenous Fgf8 protein secreted by the CoP .
To test this prediction, we performed in utero electroporations of the Sp8-engrailed construct into anterior dTel at E11.5 as above, but rather than analyzing the brains at embryonic ages, we analyzed their area patterning at P7 using serotonin immunostaining on tangential sections, which clearly delineate the primary sensory areas . We observed that Sp8-engrailed electroporation induces a substantial shift of area patterns toward the anterior pole of the cortex and a substantial reduction in the sizes of the primary sensory areas (n = 6 of 6, Figure 7j). Electroporation of the control engrailed domain alone does not affect area patterning (n = 5 of 5, Figure 7i). Thus, these anterior area shifts are consistent with an Sp8-engrailed mediated repression of targets of Sp8, such as Fgf8 and its downstream effectors (for example, Pea3). Thus, Sp8 is crucial for maintaining Fgf8 expression in the CoP and activation of its signaling pathway, as well as the influence of Fgf8 signaling on cortical arealization.
Changes in area patterning induced by ectopic expression of Fgf8
As a prelude to addressing the effect of ectopic expression of Sp8 on area patterning, we performed in utero electroporation of a CAG-Fgf8 expression construct into either anterior or posterior dTel at E11.5, and analyzed area patterning at P7 using serotonin immunostaining on tangential sections . These cases allowed us to compare the effect of ectopic expression of Sp8 to that of Fgf8 done in the same manner.
In brains with electroporations done at E11.5 and examined at E13.5/E14.5, about 80% of the transfection domains marked by the EGFP reporter are located at the region of dTel targeted for electroporation (anterior pole or posterior pole) (105 of 137 cases). However, because the EGFP reporter co-electroporated at E11.5 with the Sp8 constructs is no longer detectable at P7, to confirm these results, we repeated these studies using ROSA26 reporter mice, and co-electroporated the Sp8-VP16 dominant active construct or the VP16 control construct with a CAG-Cre Recombinase construct; the CAG-Cre Recombinase activates the β-gal reporter to mark the lineage of transfected cells and hence the location of the transfection domain. The findings confirm that the transfection domains, marked by β-gal labeled cells, are at positioned where the transfection was targeted, either the anterior or posterior cortical pole (Additional file 2), and the area shifts are identical to those described above
Sp8 induces area shifts that oppose those induced by Fgf8
Because both full-length and dominant active VP-16 forms of Sp8 induce Fgf8 and its signaling pathway, we expected that ectopic expression of Sp8 should have an effect on cortical arealization resembling that produced by ectopic expression of Fgf8. Therefore, we predicted that anterior electroporation of these Sp8 constructs would result in a posterior shift of cortical areas, similar to that for anterior overexpression of Fgf8  (Figure 8a). Based on similar logic, we anticipated that posterior ectopic expression of the Sp8 constructs should result in an anterior shift of cortical areas. We performed in utero electroporations of the Sp8 constructs into anterior or posterior dTel at E11.5, and analyzed area patterning at P7 using serotonin immunostaining on tangential sections .
Posterior electroporations of either the full-length Sp8 or Sp8-VP16 constructs result in a posterior shift of areas, accompanied by a substantial reduction in the size of V1, the most posterior primary sensory area (Sp8, n = 5 of 6; Sp8-VP16, n = 16 of 18; Figure 9h, l); again, control electroporations have no effect on area patterning. This finding is consistent with the anterior shift produced by anterior electroporation of Sp8. However, these area shifts induced by Sp8 (Figure 9) are the opposite of those observed when we electroporate an Fgf8 expression construct in posterior dTel, which we find produces an anterior shift in cortical areas (Figure 8).
Area shifts by Sp8 are paralleled by changes in markers of positional identity
To address whether area changes induced by Sp8 are accompanied by changes of positional identity of cortical neurons, we carried out a marker analysis using cadherin8 (cad8), RORβ, and ephrin-A5, each of which exhibit differential expression patterns that develop independent from thalamocortical axon input, indicating that they are bona fide markers of an intrinsic positional identity of cortical neurons [34, 35]. Cad8 expression was detected at P1 by whole mount in situ hybridization using digoxygenin-labeled probes, and RORβ and ephrin-A5 were detected using S35 labeled probes on sagittal sections. Expression was analyzed in P1 brains following in utero electroporation with Sp8 expression vectors or control vectors in anterior dTel or posterior dTel at E11.5.
In summary, although Sp8 can strongly induce Fgf8 and its signaling pathway, the changes in area patterning and the parallel shifts in positional identity of cortical neurons induced by ectopic expression of Sp8 oppose those induced by ectopic expression of Fgf8. These unexpected findings suggest the complexity of the role for Sp8 signaling in influencing cortical patterning. Sp8 might interfere with or counteract the influence of Fgf8 on area patterning without disturbing Fgf8 signaling. These opposing effects of Sp8 and Fgf8 may cooperate to balance A-P patterning of the cortex into areas.
Expression of the Sp8-like family relates to forebrain patterning centers
Based on these expression patterns, we focused on a functional relationship between Sp8 and Fgf8, as well as potential roles for Sp8 in area patterning of the neocortex. It is not possible to study area patterning using constitutive Sp8 knockout mice because they exhibit an early embryonic lethality and have severe neural tube closure defects resulting in a failure of the cortex to develop [9, 11]. Indeed, even the use of conditional knockout strategies to study roles for Sp8 in regulating Fgf8 independent from controlling cortical area patterning would be difficult as it would require the use of distinct Cre lines in which Cre recombinase is expressed in very restricted patterns, both spatially and temporally, to dissociate the function of Sp8 in regulating Fgf8 versus a direct role in cortical progenitor cells, and the requirement of Sp8 for neural tube closure and subsequent development of the cerebral cortex. To circumvent these obstacles, we used complementary gain of function and loss of function strategies employed through in utero electroporation of expression constructs, a technique similar to that recently used by others to show roles for Fgf8 in area patterning [18, 19].
Sp8 and Fgf8 share a reciprocal induction loop that exhibits significant specificity
We analyzed the relationship between Fgf8 and Sp8 in vivo by electroporating expression constructs into the lateral wall of the telencephalon and analyzing their influence at embryonic stages very early in cortical development, when Fgf8 is normally expressed in the CoP and exerts an influence on area patterning. We show that ectopic expression of Fgf8 in the telencephalon strongly induces Sp8. This finding is consistent with the recent observation that Sp8 expression is diminished in conditional or hypomorphic Fgf8 knockout mice .
We also found that ectopic expression of a full length Sp8 construct can strongly induce Fgf8 in dTel. To extend these findings, we show that ectopic expression in dTel of a 'dominant-active' Sp8-VP16 construct, created by the fusion of Sp8 with the transcriptional activator VP16, also robustly induces Fgf8, similar to the full length of Sp8, indicating that Sp8 is a transcriptional activator for Fgf8. We complemented these in vivo studies with an in vitro luciferase assay in transfected cell lines, and show that Sp8 strongly induces expression of an Fgf8 reporter construct in vitro. Thus Sp8 appears to directly induce Fgf8.
The reciprocal inductive loop between Sp8 and Fgf8 appears to be a specific relationship, consistent with the unique relationship in their expression patterns in the developing forebrain. The inductive effect of Fgf8 in the telencephalon is specific for Sp8, because we found that Fgf8 does not induce either Sp5 or Sp9 in vivo. Complementing these findings, we show using in utero electroporation in dTel that in vivo Sp9 only weakly induces Fgf8, and Sp5 has no detectable inductive effect. These findings are paralleled by those from the in vitro luciferase assay in which Sp9 weakly induced expression of an Fgf8 reporter construct. Interestingly, the inductive abilities of these Sp s for Fgf8 correlate with sequence homologies: Sp9 has identical btd and Zn finger domains to those of Sp8, whereas those of Sp5 are distinct .
In addition, the Sp8/Fgf8 inductive loop has specificity within the Fgf family, as indicated by our finding that Sp8 does not induce Fgf17, which is closely related to Fgf8 and is also expressed within the CoP. Thus, the reciprocal induction between Sp8 and Fgf8 exhibits significant specificity for one another. Such unique pairings of reciprocal induction between Sp8-like family members and forebrain patterning centers may also be utilized by Sp9 and Shh in ventral telencephalon and by Sp5 with wnts and Bmps in the cortical hem. This latter possibility is consistent with the report that the zebrafish homologue of Sp5 is induced by Wnt/β-catenin signaling .
Distinct roles for Sp8 in initiation and maintenance of Fgf8 expression in the CoP
The initiation and maintenance of Fgf8 expression in the CoP are important phenomena for the regulation of cortical area patterning. Our findings show that ectopic Sp8 expression is sufficient to induce Fgf8, but Sp8 does not appear to be required for the normal initiation of Fgf8 expression in the ANR/CoP. Analysis of Sp8 knockout mice show that Fgf8 expression is initiated in the ANR, although its subsequent expression in the ANR, or later in the CoP, is not described . Our findings, however, that Fgf8 expression is diminished when we electroporate a dominant-negative Sp8-engrailed construct into the CoP indicate that Sp8 is required to maintain Fgf8 expression in the CoP. This interpretation is also consistent with our expression data revealing that Fgf8 expression in the CoP diminishes at a stage shortly after Sp8 expression is no longer detected in the CoP but is enhanced outside of it within the cortical VZ. Thus, we conclude that although Sp8 is sufficient to induce Fgf8, it is not normally required for Fgf8 induction in the ANR/CoP. However, Sp8 is required for the maintenance of Fgf8 expression in the CoP. This scenario is similar to conclusions derived from observations of limb development in Sp8 knockout mice that Fgf8 expression is initiated but not maintained .
Sp8 induction of Fgf8 is limited to the CoP by Emx2 repression
A related important issue is the mechanism by which Fgf8 expression is limited to the CoP even though Sp8, which can robustly induce Fgf8 in the telencephalon outside of the CoP, is more broadly expressed in dTel. One possibility is that other factors uniquely present in the CoP are required to cooperate with Sp8 to induce Fgf8. Another possibility, and not mutually exclusive, is that a repressor of Sp8 induction of Fgf8 is present in the dTel but not the CoP. Multiple lines of evidence indicate that this repressor is Emx2, a homeodomain transcription factor that controls area patterning.
This interpretation is consistent with the report that the domain of Fgf8 expression expands in dTel beyond the CoP in Emx2 mutants . Furthermore, our findings provide an explanation for the report that ectopic expression of Emx2 constructs electroporated into the CoP of forebrain slices can repress Fgf8 expression when done at E11.5 embryonic stages but not at E13.5 . This temporal change in the ability of Emx2 ectopically expressed in the CoP to repress Fgf8 expression coincides with the early presence then downregulation of Sp8 expression in the CoP (present study). In conclusion, our findings indicate that the report that Emx2 both limits Fgf8 expression to the CoP and represses Fgf8 in the CoP by Emx2  are due to Emx2 repression of Sp8 induction of Fgf8 (Figure 12).
Ectopic Sp8 expression affects cortical area patterning in an opposing manner to Fgf8
We show that ectopic expression of Sp8 in either anterior or posterior dTel at embryonic stages early in cortical development has a strong effect on cortical area patterning, both the size and positioning of areas defined by serotonin staining, as well as the expression patterns of areal markers. The timing of our electroporations, and the postnatal stages of our analyses of area patterning are similar to those used by Fukuchi-Shimogori and Grove [18, 19] to study the role of Fgf8 in area patterning. They showed that anterior ectopic expression of Fgf8 causes posterior area shifts, a finding that we have confirmed. Because anterior ectopic expression of Sp8 induces Fgf8, we expected that anterior ectopic expression of Sp8 would exhibit the same phenotype as anterior ectopic expression of Fgf8. Surprisingly, though, we found that anterior ectopic expression of Sp8 results in an anterior shift of areas, the opposite effect reported for Fgf8. We also found that posterior ectopic expression of Fgf8 and Sp8 have opposing effects on area patterning: posterior ectopic expression of Sp8 results in a posterior shift of areas, whereas Fgf8 results in an opposing anterior shift. Since the same ectopic expression of Sp8 not only induces Fgf8, but also induces TFs of the ETS family that are downstream factors of Fgf8 signaling, including Pea3, Erm and Er81, the opposing effects of ectopic expression of Sp8 versus Fgf8 on area patterning is unlikely due to Sp8 triggering an aborted or defective Fgf8 signal cascade.
Thus, we hypothesize that Sp8 can induce a signaling pathway that influences area patterning in a manner that opposes the Fgf8 pathway and competes with or balances the effect of Fgf8 signaling on area patterning. Fgf8 signaling might be balanced delicately, so that small changes of Sp8 activity might result in opposing outputs; consistent with this suggestion are the findings that different dosages of Fgf8 result in opposite effects on cell death/survival and forebrain patterning [27, 38]. Sp8 might have dual roles in area patterning. In addition to a role in inducing Fgf8, Sp8 could have a more direct influence on the specification of the area identity of cortical progenitor cells that express it, such as that described for the TF EMX2 [20–22, 33]. Indeed, an intriguing study in Drosophila shows that ems, the homolog of vertebrate Emx2, specifies the identity of the head segment, but only when acting with btd, the functional homolog of vertebrate Sp8 . Thus, Sp8 may specify area identities in progenitors that express both it and Emx2, in opposing A-P gradients, potentially through an interaction with Emx2.
Timed pregnant ICR mice were used in accordance with Institution guidelines. ROSA26 reporter mice were obtained from Jackson Laboratories, and established homozygous males were crossed with ICR females to obtain heterozygous embryos. The day of insemination and the day of birth are designated as embryonic day 0.5 (E0.5) and postnatal day 0 (P0), respectively.
In situ hybridization
The following S35- or digoxigenin (DIG)-labeled riboprobes were used: Sp5, Sp8, Sp9, Fgf8 , Wnt2b (image clone 353765), Shh , Pea3, Er81 (RT-PCR), Erm (image clone 5033778), ephrin-A5, RORβ . In situ hybridization was done on 20 μm cryostat sections and DIG whole mount in situ hybridization was performed as described previously .
In utero electroporation and expression constructs
Expression constructs were made by fusing cDNA of chicken Fgf8 isoform b, full length Sp8, Sp8-VP16, Sp8-engrailed Sp5 or Sp9, which was then subcloned into a Bluescript SK(-) vector containing chicken β-actin promoter from pCAGGS vector and a polyA sequence of bovine growth hormone sequences.
Surgery and in utero electroporation were performed according to the method by Saito et al.  with some modifications as follows. Plasmid DNA (1 mg/ml for genes of interest and 1 mg/ml for pCAG-nlsEGFP or pCAG-nls-cre), mixed with 0.1% Fast green for visualization, was injected with a glass capillary at E11.5 into the lateral ventricle to transfect cells in the VZ of dTel. One of two alternative methods were then followed: paddle-type electrodes (Nepagene CUY21, 0.5 cm diameter, 30 V, 50 mS × 5) were used to obtain broad electroporation sites for studies of gene induction and needle type electrodes made of tungsten were used for area analysis to ensure precise control of the electroporation site (80 V, 50 mS × 3). Embryos were later harvested, fixed and processed. The survival rate of embryos was approximately 70%. Approximately 77% exhibited efficient GFP expression (n = 105/137) when analyzed at E13 and E14. Targeting accuracy of the electroporation sites was also evaluated by the site of the GFP domain at E13 or E14: accuracy was 76% for anterior injections (n = 87/115) and 82% for posterior injections (n = 18/22). Only one case of anterior injection resulted in middle expression of the A-P axis, and the rest were negative for GFP signals.
In vitro promoter assay
A 585 bp fragment, containing a 555 bp region immediately upstream of the mouse Fgf8 transcription start site and 30 bp of 5'-UTR of Fgf8, was subcloned into a promoter-less pGL3 basic vector (a firefly luciferase reporter vector; Promega, Madison, WI, USA), and transfected into C3H10T1/2 cells. Luciferase activity was measured after a 48 h transfection period as described previously .
Gel retardation assay
Full length Sp8 was subcloned into pcDNA3.1 (Invitrogen, Carlsbad, CA, USA) and incubated in reticulocyte lysates following the manufacture's protocols (Promega) for in vitro translation. For the gel retardation assay, 5 μl of lysate was incubated with 1 ng of the P32-labeled 585 bp fragment of DNA corresponding to the Fgf8 promoter region in binding buffer (20 mM Hepes-KOH, pH 8.0, 5 μg/ml poly(dI-dC), 2% polyvinyl alcohol) for 1 h on ice in the presence or absence of unlabeled fragments as indicated in the figure 6b. Binding products were separated in 4% polyacrylamide gels and analyzed by phosphoimage analyzer (Typhoon 8600, Molecular Dynamics, Sunnyvale, CA, USA).
Pull-down assay using biotinylated oligonucleotides
Biotinylated oligonucleotides containing the putative Sp8 binding site in the Fgf8 promoter (-230 to -190 from the transcription start site) were used. For control experiments, we introduced a mutation into the core region of the Sp1-recognition sequence, replacing GCC with TTT (designated as lower case t in the sequence below), based on a previous study . The sequences of the wild-type (WT) and mutated oligos (Mut) were: WT, 5' TGCCCGCCGCCCCGCCCCCGGCCGCCCCGCCCCCGCCCCG; Mut: 5', TGCCCGCCtttCCtttCCCGGCCtttCCtttCCCtttCCG.
A pair of complementary oligonucleotides was annealed and used as a probe. C3H10T1/2 cells were transfected with pCAGGS or pCAGGS-HA-tagged Sp8 with Fugene6 (Roche, Indianapolis, IN, USA) according to the manufacturer's instruction. Cell lysates were harvested 48 h after transfection and each cell lysate was split in half to react with WT and Mut oligos. A biotin-streptavidin pull down assay was performed for the analysis of Sp8 binding to probes as previously described [41, 42].
Histochemistry and quantification of area size
Serotonin immunostaining and X-gal staining of tangential sections were performed as described previously . The ratio of the length of motor areas and total areas of serotonin immunostained flattened brains, or cad8 expression domains of whole mount in situ P2 brains as shown in schematic diagrams in each figure, were measured by Image J software .
We thank members of the O'Leary lab, particularly T Hamasaki, S-C Chou, Y Nakagawa, and K Yee for advice, and Berta Higgins and Haydee Gutierrez and Henry Juguilon for technical help. We are grateful to P Soriano for providing the R26R mice and T Matsui for the pull-down assay protocol. This work was supported by NIH grants R37 NS31558 and R01 NS050646 (DDMO) and R01 HD034538 (JCIB). SS was supported by fellowships from the JSPS Fellowships for Research Abroad, and Uehara Memorial Foundation.
- Dalton D, Chadwick R, McGinnis W: Expression and embryonic function of empty spiracles: a Drosophila homeo box gene with two patterning functions on the anterior-posterior axis of the embryo. Genes Dev. 1989, 3: 1940-1956. 10.1101/gad.3.12a.1940.View ArticlePubMedGoogle Scholar
- Cohen S, Jurgens G: Drosophila headlines. Trends Genet. 1991, 7: 267-272.View ArticlePubMedGoogle Scholar
- Finkelstein R, Boncinelli E: From fly head to mammalian forebrain: the story of otd and Otx. Trends Genet. 1994, 10: 310-315. 10.1016/0168-9525(94)90033-7.View ArticlePubMedGoogle Scholar
- Acampora D, Annino A, Tuorto F, Puelles E, Lucchesi W, Papalia A, Simeone A: Otx genes in the evolution of the vertebrate brain. Brain Res Bull. 2005, 66: 410-420. 10.1016/j.brainresbull.2005.02.005.View ArticlePubMedGoogle Scholar
- Cecchi C: Emx2: a gene responsible for cortical development, regionalization and area specification. Gene. 2002, 291: 1-9. 10.1016/S0378-1119(02)00623-6.View ArticlePubMedGoogle Scholar
- Wimmer EA, Jackle H, Pfeifle C, Cohen SM: A Drosophila homologue of human Sp1 is a head-specific segmentation gene. Nature. 1993, 366: 690-694. 10.1038/366690a0.View ArticlePubMedGoogle Scholar
- Bouwman P, Philipsen S: Regulation of the activity of Sp1-related transcription factors. Mol Cell Endocrinol. 2002, 195: 27-38. 10.1016/S0303-7207(02)00221-6.View ArticlePubMedGoogle Scholar
- Kawakami Y, Esteban CR, Matsui T, Rodriguez-Leon J, Kato S, Belmonte JC: Sp8 and Sp9, two closely related buttonhead-like transcription factors, regulate Fgf8 expression and limb outgrowth in vertebrate embryos. Development. 2004, 131: 4763-4774. 10.1242/dev.01331.View ArticlePubMedGoogle Scholar
- Treichel D, Schock F, Jackle H, Gruss P, Mansouri A: mBtd is required to maintain signaling during murine limb development. Genes Dev. 2003, 17: 2630-2635. 10.1101/gad.274103.PubMed CentralView ArticlePubMedGoogle Scholar
- Harrison SM, Houzelstein D, Dunwoodie SL, Beddington RS: Sp5, a new member of the Sp1 family, is dynamically expressed during development and genetically interacts with Brachyury. Dev Biol. 2000, 227: 358-372. 10.1006/dbio.2000.9878.View ArticlePubMedGoogle Scholar
- Bell SM, Schreiner CM, Waclaw RR, Campbell K, Potter SS, Scott WJ: Sp8 is crucial for limb outgrowth and neuropore closure. Proc Natl Acad Sci U S A. 2003, 100: 12195-12200. 10.1073/pnas.2134310100.PubMed CentralView ArticlePubMedGoogle Scholar
- Grove EA, Fukuchi-Shimogori T: Generating the cerebral cortical area map. Annu Rev Neurosci. 2003, 26: 355-380. 10.1146/annurev.neuro.26.041002.131137.View ArticlePubMedGoogle Scholar
- O'Leary DD, Nakagawa Y: Patterning centers, regulatory genes and extrinsic mechanisms controlling arealization of the neocortex. Curr Opin Neurobiol. 2002, 12: 14-25. 10.1016/S0959-4388(02)00285-4.View ArticlePubMedGoogle Scholar
- Crossley PH, Martinez S, Ohkubo Y, Rubenstein JL: Coordinate expression of Fgf8, Otx2, Bmp4, and Shh in the rostral prosencephalon during development of the telencephalic and optic vesicles. Neuroscience. 2001, 108: 183-206. 10.1016/S0306-4522(01)00411-0.View ArticlePubMedGoogle Scholar
- Bachler M, Neubuser A: Expression of members of the Fgf family and their receptors during midfacial development. Mech Dev. 2001, 100: 313-316. 10.1016/S0925-4773(00)00518-9.View ArticlePubMedGoogle Scholar
- Grove EA, Tole S, Limon J, Yip L, Ragsdale CW: The hem of the embryonic cerebral cortex is defined by the expression of multiple Wnt genes and is compromised in Gli3-deficient mice. Development. 1998, 125: 2315-2325.PubMedGoogle Scholar
- Furuta Y, Piston DW, Hogan BL: Bone morphogenetic proteins (BMPs) as regulators of dorsal forebrain development. Development. 1997, 124: 2203-2212.PubMedGoogle Scholar
- Fukuchi-Shimogori T, Grove EA: Neocortex patterning by the secreted signaling molecule FGF8. Science. 2001, 294: 1071-1074. 10.1126/science.1064252.View ArticlePubMedGoogle Scholar
- Fukuchi-Shimogori T, Grove EA: Emx2 patterns the neocortex by regulating FGF positional signaling. Nat Neurosci. 2003, 6: 825-831. 10.1038/nn1093.View ArticlePubMedGoogle Scholar
- Bishop KM, Goudreau G, O'Leary DD: Regulation of area identity in the mammalian neocortex by Emx2 and Pax6. Science. 2000, 288: 344-349. 10.1126/science.288.5464.344.View ArticlePubMedGoogle Scholar
- Bishop KM, Rubenstein JL, O'Leary DD: Distinct actions of Emx1, Emx2, and Pax6 in regulating the specification of areas in the developing neocortex. J Neurosci. 2002, 22: 7627-7638.PubMedGoogle Scholar
- Mallamaci A, Muzio L, Chan CH, Parnavelas J, Boncinelli E: Area identity shifts in the early cerebral cortex of Emx2-/- mutant mice. Nat Neurosci. 2000, 3: 679-686. 10.1038/76630.View ArticlePubMedGoogle Scholar
- Nakamura T, Unda F, de-Vega S, Vilaxa A, Fukumoto S, Yamada KM, Yamada Y: The Kruppel-like factor epiprofin is expressed by epithelium of developing teeth, hair follicles, and limb buds and promotes cell proliferation. J Biol Chem. 2004, 279: 626-634. 10.1074/jbc.M307502200.View ArticlePubMedGoogle Scholar
- Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer RR, de Crombrugghe B: The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell. 2002, 108: 17-29. 10.1016/S0092-8674(01)00622-5.View ArticlePubMedGoogle Scholar
- Marin O, Rubenstein JL: A long, remarkable journey: tangential migration in the telencephalon. Nat Rev Neurosci. 2001, 2: 780-790. 10.1038/35097509.View ArticlePubMedGoogle Scholar
- Waclaw RR, Allen ZJ, Bell SM, Erdelyi F, Szabo G, Potter SS, Campbell K: The zinc finger transcription factor Sp8 regulates the generation and diversity of olfactory bulb interneurons. Neuron. 2006, 49: 503-516. 10.1016/j.neuron.2006.01.018.View ArticlePubMedGoogle Scholar
- Storm EE, Garel S, Borello U, Hebert JM, Martinez S, McConnell SK, Martin GR, Rubenstein JL: Dose-dependent functions of Fgf8 in regulating telencephalic patterning centers. Development. 2006, 133: 1831-1844. 10.1242/dev.02324.View ArticlePubMedGoogle Scholar
- Schafer G, Cramer T, Suske G, Kemmner W, Wiedenmann B, Hocker M: Oxidative stress regulates vascular endothelial growth factor-A gene transcription through Sp1- and Sp3-dependent activation of two proximal GC-rich promoter elements. J Biol Chem. 2003, 278: 8190-8198. 10.1074/jbc.M211999200.View ArticlePubMedGoogle Scholar
- Kawakami Y, Rodriguez-Leon J, Koth CM, Buscher D, Itoh T, Raya A, Ng JK, Esteban CR, Takahashi S, Henrique D, Schwarz MF, Asahara H, Izpisua Belmonte JC: MKP3 mediates the cellular response to FGF8 signalling in the vertebrate limb. Nat Cell Biol. 2003, 5: 513-519. 10.1038/ncb989.View ArticlePubMedGoogle Scholar
- Shimogori T, Banuchi V, Ng HY, Strauss JB, Grove EA: Embryonic signaling centers expressing BMP, WNT and FGF proteins interact to pattern the cerebral cortex. Development. 2004, 131: 5639-5647. 10.1242/dev.01428.View ArticlePubMedGoogle Scholar
- Raible F, Brand M: Tight transcriptional control of the ETS domain factors Erm and Pea3 by Fgf signaling during early zebrafish development. Mech Dev. 2001, 107: 105-117. 10.1016/S0925-4773(01)00456-7.View ArticlePubMedGoogle Scholar
- Roehl H, Nusslein-Volhard C: Zebrafish pea3 and erm are general targets of FGF8 signaling. Curr Biol. 2001, 11: 503-507. 10.1016/S0960-9822(01)00143-9.View ArticlePubMedGoogle Scholar
- Hamasaki T, Leingartner A, Ringstedt T, O'Leary DD: EMX2 regulates sizes and positioning of the primary sensory and motor areas in neocortex by direct specification of cortical progenitors. Neuron. 2004, 43: 359-372. 10.1016/j.neuron.2004.07.016.View ArticlePubMedGoogle Scholar
- Nakagawa Y, Johnson JE, O'Leary DD: Graded and areal expression patterns of regulatory genes and cadherins in embryonic neocortex independent of thalamocortical input. J Neurosci. 1999, 19: 10877-10885.PubMedGoogle Scholar
- Miyashita-Lin EM, Hevner R, Wassarman KM, Martinez S, Rubenstein JL: Early neocortical regionalization in the absence of thalamic innervation. Science. 1999, 285: 906-909. 10.1126/science.285.5429.906.View ArticlePubMedGoogle Scholar
- Weidinger G, Thorpe CJ, Wuennenberg-Stapleton K, Ngai J, Moon RT: The Sp1-related transcription factors sp5 and sp5-like act downstream of Wnt/beta-catenin signaling in mesoderm and neuroectoderm patterning. Curr Biol. 2005, 15: 489-500. 10.1016/j.cub.2005.01.041.View ArticlePubMedGoogle Scholar
- Schock F, Reischl J, Wimmer E, Taubert H, Purnell BA, Jackle H: Phenotypic suppression of empty spiracles is prevented by buttonhead. Nature. 2000, 405: 351-354. 10.1038/35012620.View ArticlePubMedGoogle Scholar
- Storm EE, Rubenstein JL, Martin GR: Dosage of Fgf8 determines whether cell survival is positively or negatively regulated in the developing forebrain. Proc Natl Acad Sci U S A. 2003, 100: 1757-1762. 10.1073/pnas.0337736100.PubMed CentralView ArticlePubMedGoogle Scholar
- Tsukui T, Capdevila J, Tamura K, Ruiz-Lozano P, Rodriguez-Esteban C, Yonei-Tamura S, Magallon J, Chandraratna RA, Chien K, Blumberg B, Evans RM, Belmonte JC: Multiple left-right asymmetry defects in Shh(-/-) mutant mice unveil a convergence of the shh and retinoic acid pathways in the control of Lefty-1. Proc Natl Acad Sci U S A. 1999, 96: 11376-11381. 10.1073/pnas.96.20.11376.PubMed CentralView ArticlePubMedGoogle Scholar
- Saito T, Nakatsuji N: Efficient gene transfer into the embryonic mouse brain using in vivo electroporation. Dev Biol. 2001, 240: 237-246. 10.1006/dbio.2001.0439.View ArticlePubMedGoogle Scholar
- Suzuki A, Raya A, Kawakami Y, Morita M, Matsui T, Nakashima K, Gage FH, Rodriguez-Esteban C, Izpisua Belmonte JC: Nanog binds to Smad1 and blocks bone morphogenetic protein-induced differentiation of embryonic stem cells. Proc Natl Acad Sci U S A. 2006, 103: 10294-10299. 10.1073/pnas.0506945103.PubMed CentralView ArticlePubMedGoogle Scholar
- Fan W, Jin S, Tong T, Zhao H, Fan F, Antinore MJ, Rajasekaran B, Wu M, Zhan Q: BRCA1 regulates GADD45 through its interactions with the OCT-1 and CAAT motifs. J Biol Chem. 2002, 277: 8061-8067. 10.1074/jbc.M110225200.View ArticlePubMedGoogle Scholar
- page NIH: .Google Scholar
- Lopez-Rovira T, Chalaux E, Massague J, Rosa JL, Ventura F: Direct binding of Smad1 and Smad4 to two distinct motifs mediates bone morphogenetic protein-specific transcriptional activation of Id1 gene. J Biol Chem. 2002, 277: 3176-3185. 10.1074/jbc.M106826200.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.